Anthropic claims that China's Alibaba used 25,000 fake accounts and 28.8 million exchanges to illicitly 'distill' its Claude model — violations occurred from April to June 2026
AI tech giant Alibaba, often considered as the Amazon of China, is being accused by Anthropic for using Claude to train its AI models. The American AI startup claimed that it traced over 25,000 accounts to operators that were affiliated with Alibaba and Alibaba Qwen.
The accusation comes at a time of escalating geopolitical tensions and increased competition in the AI sector, highlighting the aggressive tactics employed in the race for AI dominance.
This incident underscores the critical issues of intellectual property, data security, and ethical AI development, revealing strategic risks for companies developing foundational models.
The incident introduces heightened scrutiny on cross-border AI development and the potential for increased legal battles and regulatory discussions regarding AI model 'distillation' and intellectual property theft.
- · Anthropic
- · OpenAI
- · AI IP attorneys
- · Alibaba
- · Alibaba Qwen
- · Chinese AI companies
- · AI collaboration
Anthropic will likely pursue legal action against Alibaba.
Other AI companies will implement more stringent safeguards to prevent unauthorized model extraction or 'distillation'.
Governments may begin to regulate the acceptable methods for training AI models, especially those using publicly available APIs or models from competitors.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at Tom's Hardware